Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Cross-domain clusterability evaluation for cross-guided data clustering based on alignment between data domains

a clustering and data technology, applied in relational databases, web data retrieval, instruments, etc., can solve problems such as clusters that are not useful to human users in devising text analytics solutions

Inactive Publication Date: 2011-07-07
KYNDRYL INC
View PDF3 Cites 49 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Due to lack of guidance in clustering data in the target domain, conventional k-means data clustering often results in clusters that are not useful to human users in devising text analytics solutions.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Cross-domain clusterability evaluation for cross-guided data clustering based on alignment between data domains
  • Cross-domain clusterability evaluation for cross-guided data clustering based on alignment between data domains
  • Cross-domain clusterability evaluation for cross-guided data clustering based on alignment between data domains

Examples

Experimental program
Comparison scheme
Effect test

first embodiment

[0070]In step 365, the CGC finds a best match over the cross-domain similarity graph Gx that is a complete bipartite graph. Consequently, the every target cluster is aligned to one source partition, even in cases of the target cluster is dissimilar to the source partition.

second embodiment

[0071]In step 365, the CGC process disregards all edges in the match with weights below some threshold such that a target cluster is not matched to a dissimilar source partition.

third embodiment

[0072]In step 365, the CGC process defines δx( Cit, Cjs) for Dx( ) as a match weight, which is 1 if Cit is aligned with Cjs, and 0 otherwise. In this embodiment, the weaker the match with the source, the lower is the penalty for divergence from that source centroid.

[0073]After performing step 365, the CGC process proceeds with step 370.

[0074]In step 370, the CGC process updates the target centroids according to a cross-domain update rule to align the target centroids with the source centroids.

[0075]The cross-domain update rule is formulated based on assumptions that all target data items correspond to a target centroid and that the target centroid does not necessarily correspond to a source centroid. The cross-domain update rules are formulated to re-estimate respective target centroid by minimizing divergence of target clusters based on target data items and the source centroids Cs. The cross-domain update rules respective to each target centroid Cit of the target centroids Ct are ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

A system and associated method for evaluating cross-domain clusterability upon a target domain and a source domain. The cross-domain clusterability is calculated as a linear combination of a target clusterability and a source-target pair matchability, by use of a trade-off parameter that determines relative contribution of the target clusterability and the source-target pair matchability. The target clusterability quantifies how clusterable the target domain is. The source-target pair matchability is calculated as an average of a target-side matchability and a source-side matchability, which quantifies how well target centroids of the target domain are aligned with the source centroids and how well source centroids of the source domain are aligned with the target centroids, respectively.

Description

CROSS-REFERENCE TO RELATED APPLICATION[0001]This invention is related to U.S. patent application Ser. No. ______ (Attorney Docket No. IN920090070US1) entitled “CROSS-GUIDED DATA CLUSTERING BASED ON ALIGNMENT BETWEEN DATA DOMAINS”, filed on ______.BACKGROUND OF THE INVENTION[0002]The present invention discloses a system and associated method for data clustering of a target domain that is guided by relevant data clustering of a source domain, and for evaluating cross-domain clusterability of target domain data set and source domain data set. Conventional k-means data clustering generates clusters based only on intrinsic nature of data in the target domain. Due to lack of guidance in clustering data in the target domain, conventional k-means data clustering often results in clusters that are not useful to human users in devising text analytics solutions.BRIEF SUMMARY[0003]According to one embodiment of the present invention, a method for evaluating cross-domain clusterability upon a ta...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06F17/30
CPCG06F17/30598G06F17/3071G06F16/285G06F16/355G06F16/951
Inventor ACHTERMANN, JEFFREY M.BHATTACHARYA, INDRAJITENGLISH, JR., KEVIN W.GODBOLE, SHANTANU R.JOSHI, SACHINDRASRINIVASAN, ASHWINVERMA, ASHISH
Owner KYNDRYL INC
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com